LET YOUR AGENT FIND THE EDGE
A native macOS app where your AI agent loads your data, builds and backtests strategies, stress-tests them out-of-sample, and trains reinforcement-learning agents — live, in front of you. Connect Claude, Cursor, or any MCP agent.
One app · your agent · the whole research loop
The whole research loop — automated
Your agent doesn't just generate a strategy. It backtests it, proves it out-of-sample, learns from the market, and shows you what actually holds up.
Your agent drives it
Connect Claude, Cursor, or any MCP agent. 30 tools let it load data, build strategies, backtest, validate and train RL — autonomously.
Watch it live
Every step renders in a native app: strategies, equity curves, training reward curves. You see the agent think — not a black box.
Proof, not curve-fitting
Walk-forward and Monte-Carlo on every run. A robustness verdict tells you what holds up out-of-sample vs. what's overfit.
Reinforcement learning, built in
Train DQN trading agents (single or portfolio) in pure Rust — live reward curve, IS/OOS eval, then ask the model for the current signal.
Curate hundreds of runs
Every backtest and model is auto-archived. Pin the winners, delete the rest, compare them side by side — built for running hundreds.
Rust engine underneath
Millions of bars/sec, intrabar precision, 10GB+ datasets via mmap, 30+ institutional metrics (Sharpe, Sortino, VaR, CVaR).
Fast enough to
explore everything
A Rust engine doing millions of bars/sec is what lets the agent sweep thousands of strategies and train RL agents while you watch.
*Benchmark based on BTCUSDT 1m full event simulation.
RLX vs Alternatives
Detailed feature breakdown for quantitative researchers.
| Core Capabilities | RLX (Rust) | VectorBT | Backtrader |
|---|---|---|---|
| AI-Agent Tools | ✅ Native Prompt | ❌ | ❌ |
| Parallel Grid Search | ✅ Rayon/CPU | Partial | ❌ |
| Intrabar Simulation | ✅ Accurate | ❌ Vectorized | ❌ Basic |
| Institutional Metrics | 30+ | ~15 | ~10 |
| RL Environment | ✅ Integrated | ❌ | ❌ |
| Zero-Copy Data | ✅ PyO3/Numpy | ✅ | ❌ |
You give the idea. The agent does the work.
Connect your agent over MCP and tell it what to research. It runs the full loop and surfaces results in the app — you stay in the loop, not in the weeds.
Build & backtest
The agent inspects your data, drafts a strategy, validates the rules, and runs the backtest — every run archived as a report.
Walk-forward & Monte-Carlo
Out-of-sample validation and bootstrapped risk-of-ruin tell you whether the edge is real or just curve-fit to the past.
Train an RL agent
Spin up a reinforcement-learning trader that learns from the market — watch the reward curve climb and check it out-of-sample.
Ask for the forecast
“What's the call right now?” The trained model runs on the latest window and returns a long / short / flat signal in real time.
Connect your agent in two minutes
Open the app, point Claude Desktop or Cursor at its MCP endpoint, and start a conversation. No code, no notebooks — just tell the agent what to test.
- ✓ Works with any MCP agent (Claude, Cursor, …)
- ✓ 30 tools — backtest, validate, optimize, train RL, predict
- ✓ Everything the agent does shows up live in the app
Stop guessing. Let your agent prove it.
Download the Mac app, connect your agent, and run your first researched, out-of-sample-validated strategy today.